Training-a-deep-learning-model-on-the-noisy-CIFAR-dataset

This repository contains the code of training deep learning model on a noisy CIFAR dataset. The noisy dataset is generated after adding a Gaussian noise to each image. In what follows, we describe each file of this repository:

mainCifarNormalNoise.py: This file contains the main function for: 1) pre-processing the CIFAR dataset, i.e., normalizing the dataset and adding Normal noise 2) training the learning model located in the file cifarNet.py

cifarNet.py : This file contains the designed learning model TransformNormal.py : This file contains the transform for adding the Normal noise to the original dataset before starting the training